Python的pandas库实战进行一个数据处理的工作

来源：互联网发布：python 运行效率编辑：程序博客网时间：2024/06/06 04:20

下面进行一个目标处理的步骤：将对应满足要求的数据找出来进行处理。

在Excel中完全可以进行但是为了熟悉下pandas中数据框的用法，这里就花点时间试验下；
图片的格式在下方：

主函数：

main.py

import setDF2import reimport numpy as npimport pandas as pd #在data1中找出我们需要的词并输出它们的参数；准备到下次分析def fuzzyfinder(user_input, collection):        suggestions = []        pattern = '.*?'.join(user_input)    # Converts 'djm' to 'd.*?j.*?m'        regex = re.compile(pattern)         # Compiles a regex.        for item in collection:            match = regex.search(item)      # Checks if the current item matches the regex.            if match:                suggestions.append((len(match.group()), match.start(), item))        return [x for _, _, x in sorted(suggestions)]#去掉 “/n”def remove_n(l):    for i in range(len(l)):        l[i] = l[i].split('\n')[0]    return l#往一个集合里面添加一个列表里面的all元素(element)def add_all(c,s):    for e in c:        s.add(e)    return s#传递进来一个词表，返回匹配的字符串表def returnAllword(als):    set_kw = remove_n(open('C:\\Users\\Administrator\\Desktop\\word.txt','r+').readlines())    s = set()       for string in set_kw:        collection = fuzzyfinder(string,als)        s = add_all(collection,s)    al = list(s)    return al#对字符串进行二次处理，里面的字符串元素必须都是来自我们要求的字符def exchange2(l):    set_kw = remove_n(open('C:\\Users\\Administrator\\Desktop\\word.txt','r+').readlines())    aal = []        s_e = set(' ')        for st in set_kw:        s_e = add_all(list(st),s_e)    for e in l:        if(set(e) & s_e == set(e)):            aal.append(e)    return aal#已知搜索词，提取数据框中的对应数据def returnListIndex(bl):    list_all = data1.搜索词    list_index = []        for i in range(len(data1)):        if(list_all[i] in bl):            list_index.append(str(i))    return list_index'''step1: 500关键词中寻找搜索词对应的搜索词和我们对应的词条有关的词'''file = 'F:\\By\\August\\160816\\热搜探究\\0816_ws1.csv'data1 = setDF2.setDF2(file)bl = exchange2(returnAllword(data1.搜索词))list_index = returnListIndex(bl)da = np.array(bl)da.shape = len(da),1df = pd.DataFrame(da,index = da,columns = ['条件词'])data2 = pd.DataFrame(data1,index = list_index)''' step2:选取商城点击率较高 且 搜索人气>200的椅子//点击率'''re_index = []for i in np.arange(1,len(data2)):    swap = pd.DataFrame(data1,index = [data1.index[i]])    if((float(swap.搜索人气)> 200) & (float(swap.商城点击占比) > 0.40) & (float(swap.直通车参考价) < 2.57)):        re_index.append(str(i))    else:        passddv = pd.DataFrame(data1,index = re_index)print (ddv)               #print()满足条件的所有df中的关键词'''step3:将目标写出到本地'''ddv.to_csv('C:\\Users\\Administrator\\Desktop\\result_word.csv')

辅助函数setDF2.py

#等同于pandas.read_csvimport pandas as pd  import numpy as np    def strToD(x):      str1 = x.split('\n')[0]      return str1     def setDF2(file):         strings = open(file,'r+').readlines()      open(file,'r+').close()      names = [];      data = []      columes = [];      for string1 in strings[1:len(strings)]:          hang = string1.split(',')          for element in np.arange(0,len(hang)):              hang[element] = strToD(hang[element])          if(string1 == strings[1]):              columes = string1.split(',')[1:len(string1)]            columes[len(columes) - 1] =  strToD(columes[len(columes) - 1])        else:              data.extend(hang[1:len(hang)])              names.append(hang[0])    dd = np.array(data)      dd.shape = len(names),len(columes)      df = pd.DataFrame(dd,names,columes)          return df

ps:那个桌面文档的TXT就是根据特征选的关键字了；；

0 0